Title: Agencies Clarify NASH Endpoints—But Don’t Harmonize

Author: Peter G. Traber, MD

The FDA and EMA have recently published very similar clinical trial endpoints for pre-cirrhotic NASH, with one very important difference. And that one difference has important implications in Europe for the ongoing phase 3 clinical trials in pre-cirrhotic NASH.

Two recently released draft guidance documents, one from the FDA (“Noncirrhotic Nonalcoholic Steatohepatitis With Liver Fibrosis: Developing Drugs for Treatment”) and one from EMA ( “Reflection paper on regulatory requirements for the development of medicinal products for chronic non-infectious liver diseases (PBC, PSC, NASH)”), provide important industry guidance on clinical trial endpoints for clinical trials in patients with pre-cirrhotic NASH. The EMA document goes further in providing preliminary comments on NASH cirrhosis, PBC, and PSC.

Phase 3 Efficacy Endpoints

There appears to be general agreement between the Agencies on the composite clinical outcome endpoint for final market approval, although it is acknowledged that surrogate (or intermediate) endpoints are desirable for accelerated approval. It is in the use of surrogate endpoints for accelerated approval where the FDA and EMA diverge in opinion and guidance.

FDA Perspective

As taken directly from the FDA draft guidance for pre-cirrhotic NASH, “The FDA recommends sponsors consider the following liver histological improvements as endpoints reasonably likely to predict clinical benefit to support accelerated approval under the regulations:

  • Resolution of steatohepatitis on overall histopathological reading and no worsening of liver fibrosis on NASH CRN fibrosis score. Resolution of steatohepatitis is defined as absent fatty liver disease or isolated or simple steatosis without steatohepatitis and a NAS score of 0–1 for inflammation, 0 for ballooning, and any value for steatosis;


  • Improvement in liver fibrosis greater than or equal to one stage (NASH CRN fibrosis score) and no worsening of steatohepatitis (defined as no increase in NAS for ballooning, inflammation, or steatosis);


  • Both resolution of steatohepatitis and improvement in fibrosis (as defined above)”

EMA Perspective

The definition of the intermediate endpoints, in the lexicon of the EMA (which is the same as endpoints reasonably likely to predict clinical benefit to support accelerated approval by the FDA), are nearly identical to the FDA.

Again, taken directly from the EMA text, “Acceptable intermediate endpoints would consist of two composite endpoints to be evaluated at the individual patient level:

  • The resolution of NASH – with the presence of any grade of steatosis, no ballooning, and only minimal (grade 1) lobular inflammation and – at the same time – no worsening of the stage of fibrosis.
  • The improvement of fibrosis by at least 1 stage without any worsening of NASH (no worsening of ballooning and lobular inflammation, a 1 grade change in steatosis may be acceptable).”

Importantly, however, this is where the similarity of the EMA guidance departs from that of the FDA. The next paragraph of the EMA guidance states, “Efficacy in these two composites should be demonstrated in co-primary fashion, meaning that both will have to independently demonstrate a statistically significant and clinically relevant difference to placebo. This requirement is thought to take account of the uncertainties associated with a strategy to account for the long-term outcomes later.”

The impact of the difference between the FDA and EMA guidance

The FDA guidance allows flexibility to choose the most appropriate surrogate endpoint for registrational clinical trials depending on the drug mechanism of action and previous phase 2 clinical trial results. Appropriate endpoints include either resolution of NASH, or improvement in fibrosis, or both. In stark contrast, the EMA guidance stipulates the only alternative is a co-primary endpoint that has both resolution of NASH and improvement in fibrosis.

A survey of the trial designs of the four drugs currently in phase 3 clinical trials for pre-cirrhotic NASH (obeticholic acid, elafibranor, cenicriviroc, and selonsertib) reveals that each of these four trials follow the FDA definition of an appropriate accelerated approval endpoint for drug registration. In contrast, none of the four trials would meet the criterial set out in the EMA draft guidance.

The REGENERATE trial with obeticholic acid comes closest to meeting the EMA requirements but provides an option for either of the main endpoints, but not a co-primary of both:

  • The proportion of Obeticholic Acid treated patients relative to placebo achieving at least one stage of liver fibrosis improvement with no worsening of NASH,


  • The proportion of Obeticholic Acid treated patients relative to placebo achieving NASH resolution with no worsening of liver fibrosis.

The other three trials have one or the other as primary endpoints, as per the FDA guidance, but not both as stipulated by the EMA. It should be noted that each of the trials have both endpoints included in primary and secondary endpoints, but not as a co-primary endpoint.

Harmonization Imperative

Both the FDA and EMA guidance are in draft form. Clearly, it is in the best interest of bringing new drugs to patients with NASH that the two are completely harmonized. Hopefully, industry will provide the Agencies with the appropriate feedback to achieve this goal. In my view, the FDA guidance hits the mark by allowing for drugs with different mechanisms of action to have complimentary pathways to regulatory approval.

Additional NASH Insight

To read and receive other insights and commentary on drug development, please visit the Alacrita web site and subscribe to the mailing list: /consultants/peter-g-traber/

Disclosure: From March 2011 through June 2018 the author was CEO and CMO of Galectin Therapeutics (NASDAQ: GALT), a company with a NASH development asset.

For more information on how we can assist your biotech or pharmaceutical project, please reach out.

.custom-post-pagination {display: none;}